EP3825191B1

EP3825191B1 - Vehicle sideslip angle estimation system and method

Info

Publication number: EP3825191B1
Application number: EP20208182.4A
Authority: EP
Inventors: Mustafa Ali Arat; Kanwar Bharat Singh
Original assignee: Goodyear Tire and Rubber Co
Current assignee: Goodyear Tire and Rubber Co
Priority date: 2019-11-25
Filing date: 2020-11-17
Publication date: 2023-09-20
Anticipated expiration: 2040-11-17
Also published as: US11702084B2; CN112829761A; AU2020267207A1; BR102020023962A2; EP3825191A1; US20210155251A1

Description

Field of the Invention

The invention relates generally to vehicle and tire monitoring systems. More particularly, the invention relates to systems that measure and collect vehicle and tire parameter data. The invention is directed to a system and method for accurately and reliably estimating in real time a sideslip angle of a vehicle using measured parameters.

Background of the Invention

The state conditions of a vehicle vary based upon different driving conditions, such as the number of passengers in the vehicle, seating arrangement, driving terrain, and road conditions. Variation in such state conditions may affect the accuracy of vehicle state estimator models that are employed in vehicle control systems, including braking, traction, stability, and suspension systems, as well as tire state estimator models, which may also be used in the control systems and to predict tire wear. Real time knowledge of the vehicle state conditions thus is useful in many vehicle control systems as well as tire wear estimation.
Among the vehicle state conditions is the sideslip angle of the vehicle. The sideslip angle is the difference between the projected heading of the vehicle and its actual heading. Real-time knowledge of the sideslip angle of a vehicle is useful in many active vehicle safety applications. However, measurement of sideslip angle requires a complex and expensive sensor system that is generally cost prohibitive for many applications.
Strategies haven been employed to estimate sideslip angle based on specific models or observers which employ vehicle and/or tire parameters. However, these strategies have not been able to account for certain types of road conditions, such as banks or grades, and nonlinearities in vehicle motion, such as roll and pitch. The inability of prior art strategies to account for such road conditions and nonlinearities in vehicle motion undesirably reduces the accuracy of the resulting estimations. In addition, prior art techniques have been dependent upon the particular set of tires that the vehicle is equipped with which leads to inaccuracies when different tires are employed on the vehicle.
As a result, there is a need in the art for a system and method that accurately and reliably estimates vehicle sideslip angle and which accounts for road conditions and nonlinearities in vehicle motion.
US 6,547,343 B1 discloses a brake control system comprising vehicle slip estimation using a kinematic model. Parameters taken into account are steering wheel angle, vehicle speed, lateral acceleration and vehicle yaw rate.

Summary of the Invention

The invention relates to a system in accordance with claim 1 and to a method in accordance with claim 10.
Dependent claims refer to preferred embodiments of the invention.
According to an aspect of the invention, a vehicle sideslip estimation system includes a vehicle that in turn includes a controlled area network (CAN) bus system. Sensors are mounted on the vehicle in communication with the CAN bus system, and a kinematic model is in communication with the CAN bus system. The kinematic model receives signals from the sensors as inputs and estimates a lateral velocity of the vehicle based on the inputs. The inputs include a measured vehicle lateral acceleration and a vehicle longitudinal velocity. A compensated acceleration calculator is in communication with the CAN bus system to calculate a compensated vehicle lateral acceleration as a measure of conditions that result in a deviation of the measured vehicle lateral acceleration. A lateral acceleration calculator in communication with the CAN bus system to determine, based on the compensated vehicle lateral acceleration and the measured vehicle lateral acceleration, if a lateral acceleration error is larger than a predefined threshold. A filter is in communication with the CAN bus system to correct the estimated lateral velocity of the vehicle when the lateral acceleration error is larger than the predefined threshold. A velocity output register in communication with the CAN bus system to register the estimated lateral velocity of the vehicle when the lateral acceleration error is smaller than or equal to the predefined threshold. A sideslip calculator is in communication with the CAN bus system to calculate a sideslip angle of the vehicle in real time from the registered lateral velocity of the vehicle and the vehicle longitudinal velocity, wherein the compensated vehicle lateral acceleration is calculated as a non-linear function of the steering angle.
According to another aspect of the invention, a method for estimating a vehicle sideslip angle includes the steps of providing a vehicle that includes a controlled area network (CAN) bus system and sensors mounted on the vehicle that are in communication with the CAN bus system. Signals from the sensors, including a measured vehicle lateral acceleration and a vehicle longitudinal velocity, are input into a kinematic model which is in communication with the CAN bus system. A lateral velocity of the vehicle is estimated with the kinematic model based on the inputs. A compensated vehicle lateral acceleration, as a measure of conditions that result in a deviation of the measured vehicle lateral acceleration, is calculated with a compensated acceleration calculator that is in communication with the CAN bus system. It is determined if a lateral acceleration error is larger than a predefined threshold, based on the compensated vehicle lateral acceleration and the measured vehicle lateral acceleration with a lateral acceleration calculator in communication with the CAN bus system. When the lateral acceleration error is larger than the predefined threshold, the estimated lateral velocity of the vehicle is corrected with a filter in communication with the CAN bus system. When the lateral acceleration error is smaller than or equal to the predefined threshold, the estimated lateral velocity of the vehicle is registered with a velocity output register in communication with the CAN bus system. A sideslip angle of the vehicle is calculated in real time from the registered lateral velocity of the vehicle and the vehicle longitudinal velocity with a sideslip calculator in communication with the CAN bus system, wherein the step of calculating the compensated vehicle lateral acceleration with the compensated acceleration calculator includes calculating the compensated vehicle lateral acceleration as a non-linear function of the steering wheel angle.

Brief Description of the Drawings

The invention will be described by way of example and with reference to the accompanying drawings, in which:

Figure 1 is a perspective view of an exemplary vehicle employing the vehicle sideslip estimation system and method of the present invention;
Figure 2 is a plan view of the vehicle shown in Figure 1, with certain portions of the vehicle represented by dashed lines;
Figure 3 is a schematic representation of the lateral velocity and sideslip angle of a vehicle;
Figure 4A is a first schematic representation of motion stability of a vehicle defined by the sideslip angle;
Figure 4B is a second schematic representation of motion stability of a vehicle defined by the sideslip angle;
Figure 5 is an expanded perspective view of a tire, partially in section, showing a contact patch of a tire;
Figure 6 is a schematic representation of the effect of road conditions on the vehicle sideslip angle;
Figure 7 is a flow diagram showing the vehicle sideslip estimation system of the present invention and the steps of the associated method;
Figure 8A is a first graph showing performance of a prior art model-based vehicle sideslip estimation system;
Figure 8B is a second graph showing performance of a prior art model-based vehicle sideslip estimation system;
Figure 8C is a third graph showing performance of a prior art model-based vehicle sideslip estimation system;
Figure 8D is a fourth graph showing performance of a prior art model-based vehicle sideslip estimation system;
Figure 9A is a first graph showing performance of the vehicle sideslip estimation system and method of the present invention;
Figure 9B is a second graph showing performance of the vehicle sideslip estimation system and method of the present invention;
Figure 9C is a third graph showing performance of the vehicle sideslip estimation system and method of the present invention;
Figure 9D is a fourth graph showing performance of the vehicle sideslip estimation system and method of the present invention;
Figure 9E is a fifth graph showing performance of the vehicle sideslip estimation system and method of the present invention;
Figure 9F is a sixth graph showing performance of the vehicle sideslip estimation system and method of the present invention; and
Figure 10 is a flow chart showing incorporation of the vehicle sideslip estimation system and method of the present invention into a tire wear state estimation system.

Similar numerals refer to similar parts throughout the drawings.

Definitions

"ANN" or "artificial neural network" is an adaptive tool for non-linear statistical data modeling that changes its structure based on external or internal information that flows through a network during a learning phase. ANN neural networks are non-linear statistical data modeling tools used to model complex relationships between inputs and outputs or to find patterns in data.
"Axial" and "axially" means lines or directions that are parallel to the axis of rotation of the tire.
"CAN bus" or "CAN bus system" is an abbreviation for controller area network system, which is a vehicle bus standard designed to allow microcontrollers and devices to communicate with each other within a vehicle without a host computer. CAN bus is a message-based protocol, designed specifically for automotive applications.
"Circumferential" means lines or directions extending along the perimeter of the surface of the annular tread of the tire perpendicular to the axial direction.
"Equatorial Centerplane (CP)" means the plane perpendicular to the tire's axis of rotation and passing through the center of the tread.
"Footprint" means the contact patch or area of contact created by the tire tread with a flat surface, such as the ground, as the tire rotates or rolls.
"Inboard side" means the side of the tire nearest the vehicle when the tire is mounted on a wheel and the wheel is mounted on the vehicle.
"Kalman filter" is a set of mathematical equations that implement a predictor-corrector type estimator that is optimal in the sense that it minimizes the estimated error covariancewhen some presumed conditions are met.
"Lateral" means an axial direction.
"Lateral edges" means a line tangent to the axially outermost tread contact patch or footprint of the tire as measured under normal load and tire inflation, the lines being parallel to the equatorial centerplane.
"Luenberger observer" is a state observer or estimation model. A "state observer" is a system that provide an estimate of the internal state of a given real system, from measurements of the input and output of the real system. It is typically computer-implemented and provides the basis of many practical applications.
"Net contact area" means the total area of ground contacting tread elements between the lateral edges around the entire circumference of the tread of the tire divided by the gross area of the entire tread between the lateral edges.
"Outboard side" means the side of the tire farthest away from the vehicle when the tire is mounted on a wheel and the wheel is mounted on the vehicle.
"Radial" and "radially" means directions radially toward or away from the axis of rotation of the tire.
"Slip angle" is the angle between a vehicle's direction of travel and the direction in which the front wheels are pointing. Slip angle is a measurement of the deviation between the plane of tire rotation and the direction of travel of a tire.
"Tread element" or "traction element" means a rib or a block element of the tire defined by a shape having adjacent grooves.
"Tread arc width" means the arc length of the tread of the tire as measured between the lateral edges of the tread.

Detailed Description of Example Embodiments of the Invention

A first exemplary embodiment of the vehicle sideslip estimation system of the present invention is indicated at 10 in Figures 1 through 10. With particular reference to Figures 1 and 2, a vehicle 12 is supported by tires 14. While the vehicle 12 is depicted as a passenger car, the invention is not to be so restricted. The principles of the invention find application in other vehicle categories such as commercial trucks in which vehicles may be supported by more or fewer tires than shown in Figures 1 and 2.
The vehicle 12 includes a CAN bus system 16, which is a central system that enables electronic communication with sensors 18 mounted on the vehicle and/or the tires 14, and may be a wired or a wireless system. Aspects of the vehicle sideslip estimation system 10 preferably are executed on a processor 20 that is accessible through the vehicle CAN bus 16. The CAN bus 16 enables the processor 20, and accompanying memory, to receive input of data from the sensors 18 and to interface with other electronic components, as will be described in greater detail below.
Turning to Figures 3 and 5, as the vehicle 12 travels along a path 22, it experiences a lateral velocity v_y, which is a velocity of the vehicle that is orthogonal to its motion in its heading direction x. The lateral velocity v_y generates lateral forces F_y at a contact patch 24 between the tire 14 and the road 26. The lateral forces F_y are the total lateral force acting on the vehicle 12, and are direct functions of a sideslip angle β of the vehicle 12. The sideslip angle β of the vehicle 12 may be defined as the difference between projected heading and instantaneous or actual heading of the vehicle. The sideslip angle β may also be formulized by the ratio of the lateral velocity v _y and the longitudinal velocity v_x of the vehicle 12.
The sideslip angle β is a primary state of the vehicle 12 that defines motion stability. For example, as shown in Figures 4A and 4B, the sideslip angle defines oversteer (Figure 4A) and understeer (Figure 4B) conditions of the vehicle 12. The sideslip angle β is a dominating variable in vehicle dynamics analysis and is employed in characterizing the handling performance of the vehicle 12, as well as in analyzing the response of the tire 14 in lateral motion.
However, measuring the sideslip angle β is tedious. Accurate measurements of the sideslip angle β require costly equipment, which is not feasible for production vehicles, and may also not be feasible in many cases for testing purposes. In the prior art, methods have been proposed to estimate the sideslip angle β using measurements from conventional sensors 18 on the vehicle 12, such as inertial measurement units (IMUs). Such methods include estimation of the sideslip angle β and the lateral velocity v _y of the vehicle 12 using physical or model-based approaches and statistical-based approaches. The prior art model-based approaches are based on kinematic relations, which rely on accurately measured parameters. However, such models experience inaccuracies due to the omission of parameter variations, lack of tire dependent inputs and nonlinearities. The prior art statistical-based approaches, which include machine learning, employ training or tuning, which must be performed very carefully over an extensive set of use cases. Such extensive training or tuning renders such statistical-based approaches impractical.
Essential challenges in estimating the lateral velocity v _y and the resulting sideslip angle β of the vehicle 12 lie in matching the exogenous or external inputs from the road 26 and nonlinearities in vehicle motion. Turning to Figure 6, road inputs, such as road bank or grade, are not feasible to be generalized and modeled. In addition, nonlinearities in the motion of the vehicle 12, such as roll and pitch, render mathematical models inapplicable in estimating vehicle states, including sideslip angle β.
The vehicle sideslip estimation system 10 of the present invention provides a system and accompanying method to estimate the lateral velocity v_y and the sideslip angle β of the vehicle 12 in real time, employing conventional sensors 18 that are available on the vehicle. The vehicle sideslip estimation system 10 captures the effects of unknown inputs from the road 26, as well as nonlinearities in vehicle response, based on a change in lateral acceleration a_y through operation of the vehicle 12. The vehicle sideslip estimation system 10 calculates such exogenous inputs and nonlinearities and filters them for correction of the vehicle state.
With reference to Figure 7, a flow diagram shows the system of the vehicle sideslip estimation system 10 and illustrates the steps of the accompanying method. Nomenclature is as follows:

δ :: steering wheel angle
r:: yaw rate
ax:: longitudinal acceleration
ay:: lateral acceleration
vx :: longitudinal velocity
vy :: lateral velocity
β :: sideslip angle
Kvx:: longitudinal velocity gain value
Kvy:: lateral velocity gain value

Signals from sensors 18 that are attached to the vehicle 12 are available from the CAN bus 16 and provide measured values for the steering wheel angle δ, the yaw rate r, the vehicle longitudinal acceleration a_x , the vehicle lateral acceleration a_y, and the vehicle longitudinal velocity v_x as inputs into a kinematic model 28. Preferably, the sensors 18 are conventional sensors that are available on production vehicles 12, such as body accelerometers, rate gyros, speed sensors and steering wheel angle sensors.
The kinematic model 28 is in communication with the CAN bus 16 and preferably employs an adaptive sliding mode observer (SMO) 30 to estimate the lateral velocity v_y of the vehicle 12. For example: ${\hat{v}}_{y} = f (a_{y}, r, {\hat{v}}_{x}, K_{v_{y}})$
Where the lateral velocity v_y is a function of the vehicle lateral acceleration a_y , the yaw rate r, the vehicle longitudinal velocity v_x, and the lateral velocity gain value K_vy in the observer 30.
While the longitudinal velocity v_x is obtained from the vehicle CAN bus 16 as a measured signal, it is updated for observability of the system 10 and to act as a filter on the CAN bus signal: ${\hat{v}}_{x} = f (a_{x}, r, {\hat{v}}_{y}, K_{v_{x}})$
The update of the longitudinal velocity v_x is a function of the vehicle longitudinal acceleration a_x , the yaw rate r, the last estimated lateral velocity v_y , and the longitudinal velocity gain value K_vx. The gain values K_vx and K_vy penalize the error between the measured signals and the corresponding values that are calculated by the observer 30, which enables the states of the observer, and specifically the lateral velocity v_y , to converge to the physically correct value.
The exemplary adaptive SMO 30 preferably uses kinematic relations for the lateral and longitudinal motion of the vehicle 12 in relative terms, and as mentioned above, includes an adaptive gain definition based on driver steering input and corresponding lateral acceleration for the estimation. An exemplary adaptive SMO 30 estimation is calculated as follows: $\dot{\hat{v}} = a_{y} - {rv}_{x} + (a_{y_{corr}} + K_{v_{y}} \tanh ({\tilde{a}}_{y}))$
${\dot{\hat{v}}}_{x} = a_{x} + r {\hat{v}}_{y} + (K_{v_{x}} \tanh (v_{x} - {\hat{v}}_{x}))$
The value of tanh is a hyperbolic tangent that is used to obtain the sign of the error signal without introducing discontinuity.
A compensated acceleration calculator 32 is in communication with the CAN bus 16 and is employed to account for input bias due to roll motion of the vehicle 12. More particularly, because the measured lateral acceleration a_y is typically biased due to vehicle roll motion (as shown in Figure 6), the estimated lateral states, including the estimated lateral velocity v_y , diverge. A compensated lateral acceleration â_y is calculated, which is used as a measure of the conditions that yield significant roll motion and result in deviation or bias of the measured lateral acceleration a_y obtained from the vehicle CAN bus 16.
In the compensated acceleration calculator 32: ${\hat{a}}_{y} \approx η (t) {\hat{v}}_{y}$
Where â_y is the compensated lateral acceleration, v̂_y is the estimated lateral velocity from the kinematic model 28, and η(t) is a time-varying gain value that is employed to approximate an expected planar acceleration of the vehicle 12. The compensated acceleration â_y is calculated as a nonlinear function of the steering wheel angle as follows: ${\tilde{a}}_{y} = a_{y} - [\frac{K_{1}}{1 + e^{(- K_{2} δ)}}] {\hat{v}}_{y}$
The value K₁ is a constant gain value used to scale the error level and is tuned for a given platform of the vehicle 12.
Once the compensated lateral acceleration â_y is determined, a lateral acceleration calculator 34, which is in communication with the CAN bus 16, determines if an error in lateral acceleration e_y, also referred to as a lateral acceleration error, is larger than a predefined threshold T_y. Specifically, the lateral acceleration calculator 34 defines the error in lateral acceleration e_y as the difference between the measured lateral acceleration a_y obtained from the CAN bus 16 and the compensated lateral acceleration â_y : $e_{y} : = a_{y} - {\hat{a}}_{y} > T_{y}$
To determine the predefined threshold T_y, a road friction condition estimator 36 estimates a road friction µ at a specific time t, which may be expressed as µ(t). The road friction condition estimator 36 is in communication with the CAN bus 16. An exemplary system and method for estimating the road friction µ at time t is shown and described in US-B-9,751,533 . Of course, other known systems and methods for estimating the road friction µ may be employed in the road friction condition estimator 36. The road friction condition estimator 36 ensures operation of the system 10 in real time t.
The predefined threshold T_y is then determined in a threshold calculator 38, which is in communication with the CAN bus 16. The threshold calculator 38 calculates an updated threshold value as a function of the real-time road friction µ: $T_{y} \approx f (μ)$
The calculation of the updated threshold T_y preferably is performed with a linear regression algorithm, such as: $T_{y} = a_{1} μ + a_{2}$
The linear regression algorithm employs the road friction µ as an input and is trained for a given platform of the vehicle 12. The values a₁ and a₂ are constant coefficients of the regressor, which are computed in the process of training the regressor model using historical data, including data of the lateral acceleration signal on varying road friction µ conditions for the selected vehicle platform.
The updated predefined threshold T_y is input from the threshold calculator 38 into the lateral acceleration calculator 34 to determine, as described above, if the error in lateral acceleration e_y is larger than the predefined threshold.
When the error in lateral acceleration e_y is larger than the predefined threshold T_y, the corrected lateral acceleration is modeled as a second order system of the error: ${\ddot{a}}_{y_{corr}} + τ_{1} {\dot{a}}_{y_{corr}} + τ_{2} = K_{f} {\tilde{a}}_{y}$
This equation models a filter 40 which is employed to correct the estimated lateral velocity v _y . More particularly, the filter 40 is in communication with the CAN bus 16 and is a second-order low pass filter that calculates a filtered value or corrected lateral acceleration a_ycorr : $a_{y_{corr}} = G (s) {\tilde{a}}_{y}$
where $G (s) = (\frac{K_{f}}{s^{2} + τ_{1} s + τ_{2}})$
In the filter 40, s is the Laplace operator, and K_f, τ₁ and τ₂ are constants for the filter design that are tuned for a given platform of the vehicle 12, in which K_f is static sensitivity, τ₁ is a first order time constant, and τ₂ is a second order time constant.
The filtered value a_ycorr is fed back into the kinematic model 32. It is used in summation with the observer gain K_vy and integrated into the adaptive SMO 30 to accurately estimate the lateral velocity v _y .
When the error in lateral acceleration e_y is smaller than or equal to the predefined threshold T_y, the estimated lateral velocity v _y is registered as an accurate lateral velocity in a velocity output register 42. The velocity output register 42 is in communication with the CAN bus 16, and the registered lateral velocity v _y is for the particular time t.
The registered lateral velocity v _y and the measured longitudinal velocity v_x are inputs into a sideslip calculator 44, which is in communication with the CAN bus 16 and calculates a value for the sideslip angle β at the particular time t. The sideslip calculator employs a recursive least squares algorithm to determine the sideslip angle β from the registered lateral velocity v _y and the measured longitudinal velocity v_x : $β = v_{y} / v_{x}$
The resulting value of the sideslip angle β thus is an accurate value that accounts for road inputs, such as road bank or grade, as well as nonlinearities in the motion of the vehicle 12, such as roll and pitch.
Turning to Figures 8A through 8D, test results of a prior art model-based approach are shown, in which sideslip angle β is plotted as a function of time t in vehicle maneuvers. As indicated by the measured values 46 as compared to the estimated values 48, the model-based approach performs reasonably well under consistent conditions, which are the maneuvers shown in Figures 8A and 8B. However, as the parameters vary due to a tire change, Figure 8C, or model nonlinearities, Figure 8D, the measured values 46 and the estimated values diverge, indicating inaccuracies in the model-based approach.
With reference to Figures 9A through 9F, test results of the vehicle sideslip estimation system 10 are shown in which sideslip angle β is plotted as a function of time t in vehicle maneuvers. As indicated in by the measured values 50 as compared to the estimated values 52, the sideslip estimation system 10 performs well when parameters vary due to a tire change, Figures 9A through 9D, or model nonlinearities, Figures 9E and 9F. Such results show the accuracy of the vehicle sideslip estimation system 10.
Referring to Figure 10, the accurate sideslip angle β determined by the vehicle sideslip estimation system 10 may be communicated to the vehicle CAN bus 16 to be employed in vehicle control systems, such as stability and handling systems, as well as tire state estimator models that predict tire wear. For example, the sideslip angle β may be used to calculate a tire slip angle 54, including a specific front tire slip angle 54F and a specific rear tire slip angle 54R. These slip angles 54F and 54R may be employed as an input in estimation or prediction systems, such as those shown and described US-B-9,428,013 and US-B-9,873,293 . Such estimation or prediction systems may thus employ the sideslip angle β determined by the vehicle sideslip estimation system 10 to generate a robust and reliable tire wear state or condition estimate 56.
In this manner, the vehicle sideslip estimation system 10 accurately and reliably accurately and reliably estimates a vehicle sideslip angle β that accounts for road conditions and nonlinearities in vehicle motion, including road bank angle, road friction and the like, as well as a change in tires 14. Road inputs and un-modeled nonlinearities are matched by means of the error between an expected lateral acceleration, which is calculated based on a time-varying gain and the measured lateral acceleration which includes the aforementioned effects.
In addition, the vehicle sideslip estimation system 10 is tire agnostic, and thus provides a robust and accurate system even when different tires 14 are mounted on the vehicle 12.
The present invention also includes a method for estimating a sideslip angle β of a vehicle 12. The method includes steps in accordance with the description that is presented above and shown in Figures 1 through 10.
It is to be understood that modifications are possible within the scope of the invention, as defined by the appended claims.

Claims

A vehicle sideslip estimation system (10) comprising:
a vehicle (12) including a controlled area network (CAN) bus system (16);

sensors (18) mounted on the vehicle (12) in communication with the CAN bus system (16);

a kinematic model (28) in communication with the CAN bus system (16), the kinematic model (28) being configured for receiving signals from the sensors (18) as inputs, wherein the kinematic model (28) is configured for estimating a lateral velocity of the vehicle (12) based on the inputs, the inputs including a measured vehicle lateral acceleration and a vehicle longitudinal velocity;

a compensated acceleration calculator (32) in communication with the CAN bus system (16) to calculate a compensated vehicle lateral acceleration as a measure of conditions that result in a deviation of the measured vehicle lateral acceleration;

a lateral acceleration calculator (34) in communication with the CAN bus system (16) to determine, based on the compensated vehicle lateral acceleration and the measured vehicle lateral acceleration, if a lateral acceleration error is larger than a predefined threshold;

a filter (40) in communication with the CAN bus system (16) to correct the estimated lateral velocity of the vehicle (12) when the lateral acceleration error is larger than the predefined threshold;

a velocity output register (42) in communication with the CAN bus system (16) to register the estimated lateral velocity of the vehicle (12) when the lateral acceleration error is smaller than or equal to the predefined threshold; and

a sideslip calculator (44) in communication with the CAN bus system (16) to calculate a sideslip angle of the vehicle (12) in real time from the registered lateral velocity of the vehicle and the vehicle longitudinal velocity;

wherein the compensated vehicle lateral acceleration is calculated as a nonlinear function of the steering wheel angle.
The vehicle sideslip estimation system of claim 1, wherein the inputs into the kinematic model (28) from the sensors (18) further comprise a steering wheel angle, a yaw rate, and a vehicle longitudinal acceleration.
The vehicle sideslip estimation system of claim 1 or 2, wherein the kinematic model (28) is configured to employ an adaptive sliding mode observer (30) to estimate the lateral velocity of the vehicle (12), the lateral velocity preferably being a function of the measured vehicle lateral acceleration, the yaw rate, the vehicle longitudinal velocity, and a lateral velocity gain value.
The vehicle sideslip estimation system of at least one of the previous claims, wherein the longitudinal velocity of the vehicle (12) is updated for observability of the system (10) as a function of the vehicle longitudinal acceleration, the yaw rate, a last estimated lateral velocity, and a longitudinal velocity gain value.
The vehicle sideslip estimation system of at least one of the previous claims, wherein the lateral acceleration error is defined as the difference between the measured vehicle lateral acceleration and the compensated vehicle lateral acceleration.
The vehicle sideslip estimation system of at least one of the previous claims, wherein a road friction condition estimator (36) is configured to estimate a road friction in real time, and a threshold calculator (38) is configured to update the predefined threshold as a function of the real time road friction, preferably with a linear regression algorithm.
The vehicle sideslip estimation system of at least one of the previous claims, wherein the filter (40) is a second-order low pass filter that is configured to calculate a filtered value which is input into the kinematic model (28).
The vehicle sideslip estimation system of at least one of the previous claims, wherein the sideslip calculator (44) is configured to employ a recursive least squares algorithm to calculate the sideslip angle of the vehicle (12) from the registered lateral velocity of the vehicle and the vehicle longitudinal velocity.
The vehicle sideslip estimation system of at least one of the previous claims, further comprising a tire state estimator model (56) that is configured to receive the sideslip angle of the vehicle to determine a tire wear state.
A method for estimating a vehicle sideslip angle, the method comprising the steps of:
providing a vehicle (12) that includes a controlled area network (CAN) bus system (16) and sensors (20) mounted on the vehicle (12) in communication with the CAN bus system (16);

inputting into a kinematic model (28), which is in communication with the CAN bus system (16), signals from the sensors (18), including a measured vehicle lateral acceleration and a vehicle longitudinal velocity;

estimating with the kinematic model (28) a lateral velocity of the vehicle (12) based on the inputs;

calculating with a compensated acceleration calculator (32), which is in communication with the CAN bus system (16), a compensated vehicle lateral acceleration as a measure of conditions that result in a deviation of the measured vehicle lateral acceleration;

determining with a lateral acceleration calculator (34), which is in communication with the CAN bus system (16), if a lateral acceleration error is larger than a predefined threshold based on the compensated vehicle lateral acceleration and the measured vehicle lateral acceleration;

correcting with a filter (40), which is in communication with the CAN bus system (16), the estimated lateral velocity of the vehicle (12) when the lateral acceleration error is larger than the predefined threshold;

registering with a velocity output register (42), which is in communication with the CAN bus system (16), the estimated lateral velocity of the vehicle (12) when the lateral acceleration error is smaller than or equal to the predefined threshold; and

calculating with a sideslip calculator (44), which is in communication with the CAN bus system (16), a sideslip angle of the vehicle in real time from the registered lateral velocity of the vehicle (12) and the vehicle longitudinal velocity;

wherein the step of calculating the compensated vehicle lateral acceleration with the compensated acceleration calculator (32) includes calculating the compensated vehicle lateral acceleration as a nonlinear function of the steering wheel angle.
The method for estimating a vehicle sideslip angle of claim 10, wherein the step of inputting signals from the sensors (18) into the kinematic model (28) includes inputting a steering wheel angle, a yaw rate, and a vehicle longitudinal acceleration; and/or wherein the step of estimating the lateral velocity of the vehicle (12) with the kinematic model (28) includes employing an adaptive sliding mode observer (30) to estimate the lateral velocity of the vehicle (12).
The method for estimating a vehicle sideslip angle of at least one of the claims 10 to 11, wherein the step of determining if the lateral acceleration error is larger than the predefined threshold with the lateral acceleration calculator (34) further comprises estimating a road friction in real time with a road friction condition estimator (36), and updating the predefined threshold as a function of the real time road friction with a threshold calculator (38), and, preferably, wherein the step of updating the predefined threshold as a function of the real time road friction with a threshold calculator (38) includes the use of a linear regression algorithm.
The method for estimating a vehicle sideslip angle of at least one of the claims 10 to 12, wherein the step of correcting the estimated lateral velocity of the vehicle (12) with a filter (40) includes the use of a second-order low pass filter, and inputting a calculated filtered value into the kinematic model (28); and/or wherein the step of calculating a sideslip angle of the vehicle (12) with a sideslip calculator (44) includes using a recursive least squares algorithm in the sideslip calculator (44).